This paper reviews some of the neurochemical and neurophysiological literature that is relevant to the hallucinatory effects of psychedelic drugs and their residual effects in Hallucinogen Persisting Perception Disorder (HPPD). A hypothesis is proposed of a possible neurophysiological mechanism of some of the characteristic visual distortions caused by these drugs and seen in HPPD. The posited mechanism involves the induction of inhibition in a few key areas in the visual system which significantly alters the normal processing of object edges in the visual scene, resulting in some stereotypical visual distortions. Suggestions are then made for possible research directions that might lend support to this hypothesis.
Lysergic Acid Diethylamide and other psychedelic drugs exhibit a wide variety of perceptual, cognitive, and emotional effects that vary greatly between different users, settings, and dosages. Despite the variation in reported effects, there are certain perceptual, especially visual, distortions that are described by a large number of users of psychedelic substances. Among the most common visual distortions are hallucinations of geometric patterns, "trails" that follow moving stimuli, false perceptions of movement in the peripheral visual fields, "halos" around objects, alterations in the apparent size of objects, and synesthesia, the experience of stimuli of one sensory modality as if it were in a different modality ("seeing music" and "hearing colors", for example). Many of these visual distortions are also present in Hallucinogen Persisting Perception Disorder (HPPD), defined by the DSM-IV (Diagnostic and Statistical Manual of Mental Disorders) as the "reexperiencing, following cessation of use...of the perceptual symptoms that were experienced while intoxicated with the hallucinogen" (DSM-IV, 1994; Abraham 1983).
This paper will attempt to synthesize the literature on the subjectively reported and empirically studied visual distortions in both the acute, short-term effects of hallucinogens and their long-term effects seen in HPPD with the literatures on the neurochemical and anatomical loci of LSD's effects and on the neurophysiology and anatomy of related visual circuitry. The end result will be a hypothesis of a possible mechanism that could underlie several of the reported visual distortions experienced by LSD users. It is important to note, however, that all three literatures that this paper draws on have far to go in fully exploring and understanding their respective subjects, which are quite complex. Furthermore, even the current knowledge of the actions of LSD implicates a plethora of neural structures and systems, far beyond the scope of this paper and its conjectures. Thus, it is quite likely that the hypotheses put forth in this paper are incorrect, and if correct, they certainly cannot explain all of the highly subjective alterations of consciousness produced by psychedelic drugs. This paper will, however, also suggest possible research directions that could support or invalidate the hypotheses it presents.
The hypothesis, based on structural similarity and some empirical evidence, that psychedelic drugs exert their effects primarily in the serotonergic (5-HT) system of the brain has retained wide acceptance for many decades (Aghajanian 1994). This hypothesis was thrown into question when it was found that indoleamine hallucinogens, such as LSD, produce physiological effects that are mediated by their binding to the 5-HT-1A receptor subtype, which are not produced by other chemical classes of hallucinogens, such as the phenethylamine, mescaline (Aghajanian 1994). It was reasoned that these two classes of drugs, which produce phenomenologically similar hallucinatory experiences, should hypothetically act through similar neurochemical mechanisms.
More recent experiments have suggested a solution to this apparent paradox. High correlations have been found between the affinity of different classes of hallucinogens for the 5-HT-2 receptor subtype and their potency in producing hallucinations (Titeler et al. 1988). Assayed using various techniques and subject species, regional distributions of 5-HT-2 receptor density and of the binding of various hallucinogens seems to correspond well (McKenna et al. 1989; Saavendra 1989; Watts et al. 1994). Structures that reliably show high concentrations of both include: parts of the limbic system and basal ganglia, piriform cortex, dentate gyrus, and, most importantly to this paper, the claustrum and layers 3 and 4 of the cerebral cortex. Further evidence comes from the reversal of certain physiological effects of psychedelics by selective 5-HT-2 antagonists (Rasmussen and Aghajanian 1986). Thus, the 5-HT hypothesis has been refined to its current form, that the hallucinogenic properties of both these classes of drugs are mediated by their effects on the postsynaptic 5-HT-2 system.
To attempt to relate the behavioral effects of a psychoactive chemical to known neural circuitry, one must first bridge the gap between these disparate areas with knowledge of the specific physiological effects of the substance on the relevant circuitry. Some recent studies have contributed significantly to our understanding of the physiological effects of serotonin at the 5-HT-2 receptor, and of the relationship between this receptor subtype and hallucinogenic drugs. These experiments are providing growing evidence that 5-HT-2 receptors in the cerebral cortex and related structures are found mostly on GABAergic (gamma-aminobutyric acid) inhibitory interneurons, which tend to inhibit the principal cells, and that hallucinogens are potent partial agonists at these synapses.
Sheldon and Aghajanian (1990) found a serotonin-mediated increase in inhibitory post-synaptic potentials (IPSPs) in the pyramidal cells (the principal cells) of piriform cortex, an olfactory structure that has been described as a phylogenetically primitive version of the neocortex. These IPSPs were blocked by concurrently applied bicuculline, a GABA antagonist, and by ritanserin, a selective 5-HT-2 antagonist, suggesting that serotonergic cells provide excitatory input to GABAergic inhibitory interneurons, which then inhibit pyramidal cells. LSD and the phenethylamine hallucinogen DOI ([+]-1-[2,5-dimethoxy-4-iodophenyl]-2-aminopropane) produce responses that are very similar to serotonin, but weaker, in piriform inhibitory interneurons (Marek and Aghajanian 1996). Though these responses suggest that psychedelics are partial agonists at these synapses, their effectiveness at very low concentrations (nanomolar for LSD) gives evidence that they are highly potent.
Similar mechanisms seem to exist in the prefrontal cerebral cortex. Anatomical studies have shown that serotonergic axons innervating monkey prefrontal cortex synapse predominantly on interneurons (Smiley and Golman-Rakic 1996). Local injection of 5-HT or the 5-HT-2 agonist (and hallucinogen) DOI increases the concentration of GABA in prefrontal cortex in a dose-dependent manner, and there is evidence that DOI preferentially targets GABA interneurons (Abi-Saab et al. 1996). Dopaminergic and serotonergic systems have been shown to produce similar inhibitory responses in prefrontal cortex, but only the latter was eliminated by a selective 5-HT-2 antagonist (Godbout, et al. 1991). Furthermore, LSD has been shown to enhance the dopaminergic mediated response, but this enhancement is blocked by a 5-HT-2 antagonist (Godbout, et al. 1991).
Therefore, it seems that in both piriform and prefrontal cortex, hallucinogens have pharmacologically specific potent excitatory effects on interneurons that inhibit pyramidal cells. Since the piriform is one of the phylogenetically oldest cortical structures, and the prefrontal one of the newest, it seems likely that similar mechanisms exist in the other cortical and cortex-related structures that are innervated by 5-HT-2 axons. This leap of logic is supported by findings of similar laminar distribution of 5-HT-2 receptors in almost all cortical areas (Lidow et al. 1989).
One area that may contain similar serotonergic systems is the claustrum, a nucleus that is buried in white matter between the putamen of the basal ganglia and the insula of the frontal neocortex. Because of its location, its anatomy and cell morphology and its extensive connections with almost all of the neocortex (and its comparatively sparse connections with subcortical structures), it has been posited that the claustrum was once a part of the cortex that somehow, in the course of evolution, broke off and became a physically separate nucleus (Sherk 1986). As previously mentioned, convergent evidence shows that there is a high concentration of 5-HT-2 receptors, and of hallucinogen binding (both indoleamines and phenethylamines) in the claustrum. Therefore, based on its anatomical and neurochemical properties, it seems likely that the claustrum is affected by psychedelics through mechanisms similar to that found in the piriform and prefrontal cortices.
The claustrum has widespread connections, usually reciprocal, with almost all areas of the cerebral cortex, including striate and extrastriate visual areas, auditory, somatosensory and motor areas, frontal, parietal, and temporal association cortex, and cingulate cortex. The claustral zones corresponding to most of these areas tend to preserve their topographic organization, especially those parts of the claustum that are connected to areas with well-defined topographies. There is much debate about how much the input regions from different cortical areas overlap, and whether individual claustrum neurons are connected to more than one cortical area. There seems to be general trend, though, that cortical areas that are connected with one another, and/or have related functions, tend to have at least partially overlapping claustral projection zones. For example, the zone that projects to primary visual cortex (V1) overlaps those that are linked to higher order visual areas. These higher order extrastriate areas also project to prefrontal areas, and accordingly, their claustral target zones also overlap. Since it has yet to be shown that individual claustral neurons project to different cortical areas, any cross-modal theory of its function is quite speculative, however.
Like all neocortex, primary visual cortex (V1) is divided into 6 layers, and each layer has somewhat different receptive field properties and connections. In most layers, V1 cells respond selectively to a line or bar of light that is swept across its receptive field at a specific orientation and direction. In layer 4, and other more superficial layers, cells also exhibit end-stopping (also called end-inhibition)--they respond well to lines up to a certain length, but longer lines inhibit their response. Layer 6 neurons have complementary properties--they respond little to short lines, but are increasingly excited by longer stimuli, and are not end-inhibited. Bolz and Gilbert (1986) have proposed a mechanism for the induction of end-inhibition in the upper layers of V1: layer 6 cells have excitatory projections to inhibitory interneurons, which inhibit cells in layer 4. A short line will effectively activate layer 4, but not layer 6, neurons, while a long stimulus will increase the response of layer 6 cells, which will then induce inhibition of layer 4 cells, and therefore they will not respond well to these long stimuli. This hypothesis is supported by the loss of end-inhibition in layer 4 when layer 6 is selectively inactivated by micro-injections of inhibitory neurotransmitter (Bolz and Gilbert 1986).
V1's projections to the claustrum arise almost exclusively from layer 6 (LeVay and Sherk 1981a), and, not surprisingly, the claustral cells they innervate also respond well to very long moving bars (Sherk and LeVay 1981). The claustrum's return projection to V1 seems to preferentially innervate layer 4 inhibitory interneurons (LeVay 1986), and, interestingly, lesioning the claustrum reduces end-inhibition in layer 4, though to a somewhat lesser degree than inactivation of V1 layer 6 (Sherk and LeVay 1983). No other measured response properties of layer 4 cells were changed by inactivation of either the claustrum or layer 6, but it is important to note, however, that the measurements used by LeVay were not exhaustive of the vast possibilities, and they were carried out on anesthetized animals, precluding the uncovering of any possible cognitively-mediated functions of the claustrum (see below).
Therefore, layer 6 of V1 and the claustrum appear to be parallel circuits for generating end-inhibition in the upper layers of V1. The purpose of having parallel mechanisms for this function is not known, however. The V1 target zone in the claustrum appears to overlap to some degree with those of higher order visual areas, which are specialized for the processing of motion, color, and form, and with areas of parietal and prefrontal cortex that have been implicated in attention, working memory, and the generation of eye movements (Sherk 1986; Baizer et al. 1993; Boussaoud et al. 1992; Selemon and Goldman-Rakic 1988). This suggests that a possible function of the cortico-claustral loop is to regulate V1 end-inhibition based on more global aspects of the visual scene (such as whether the line in a cell's receptive field defines the edge of an object, or is just part of the background) based on upcoming eye movement or expected motion of an object, or based on what one is paying attention to or actively looking for in a visual scene.
The perceptual function of end-inhibition also remains an open question. It has been suggested that pooling of the responses of end-inhibited cells allows for the perception of so-called "illusory contours" (e.g. von der Heydt and Peterhans 1989). These types of borders are referred to as illusory because there is no actual line or luminance difference defining them, yet human observers consistently perceive an edge in them. An example of this type of illusory contour would be two stacks of several horizontal lines whose ends are touching one another, but are slightly offset vertically (usually by half the distance between each line in the stack); this induces the perception of an illusory vertical edge at the border between these two stacks of lines. Since this edge is defined only by the ends of a series of lines, it is hypothesized that the mechanism that detects illusory contours is one that integrates the signals from several end-inhibited cells. It seems likely that our visual system, being quite efficient at what it does, uses these same mechanisms to aid in the perception of the (non-illusory) borders of all objects, especially in cases of edges that are difficult to accurately perceive. Our perception of any vertical object edge, for instance, may be the result of a combination of information from both cells that are selective for vertical lines and end-inhibited cells selective for horizontal lines whose ends are lined up against this vertical border in the visual field.
Traditionally, the response properties of neurons have been characterized solely by measuring their average rate of firing action potentials over a relatively long period of time. However, it is becoming increasingly acknowledged that temporal patterns in the firing of neurons may also be very important in transmitting information (e.g. Abeles 1991). For example, a group of cells that all respond to a line of a specific orientation and location would be much more likely to effectively drive a common post-synaptic cell if they all fired simultaneously (because of the summation of their post-synaptic responses) than if they all fired at different times with the same average firing rate. Thus, the signal they coded for would have a greater chance of getting propagated to higher levels of processing, and presumably, of significantly influencing one's consciousness.
Such temporal coordination could be very important for neural mechanisms that detect edges (either illusory or actual) from the aligned ends of several end-inhibited cells. Integrating the responses of perhaps thousands of end-stopped cells, and combining this with edge information provided by orthogonally oriented line-detecting cells is an immensely complex computational task that is multiplied by the almost constant motion of objects in the visual field (due to both their intrinsic motion and to our own frequent eye movements). A system that performed such a computation though passive integration of firing rates over time spans of even hundreds of milliseconds would not seem to be sufficient to support our rich and reliable perceptions of object boundaries.
Therefore, I propose that the integrity of the visual cortico-claustral loop may be necessary for precise temporal integration of edge information from end-inhibited and line-detector cell populations, and that the nature of this temporal coding may be modulated based on the expected motion of objects, eye movements, and selective attention. When this system is compromised, either because of experimental manipulations or drug interactions, information about object boundaries will be processed abnormally, resulting in perceptions that are degraded or altered in some way. Furthermore, I hypothesize that LSD and other hallucinogens excite inhibitory interneurons in both the claustrum and in layer 4 of V1, similar to their effects in other cortical regions. Recent theories and empirical evidence suggest that inhibition may play a critical role in coordinating the temporal patterns of neuronal responses to stimuli (Singer 1996). Therefore, the dysfunction of inhibition in these areas caused by psychedelics may, at least in part, mediate the visual distortions they produce. The persistence of these distortions in HPPD suggests that, in some users, hallucinogens may cause long-term neurochemical changes, such as an increase in the sensitivity or number of 5-HT-2 receptors, which result in similar effects.
One of the most commonly reported visual distortions related to both the acute effects of LSD, and its residual effects seen in Hallucinogen Persisting Perception Disorder is "trailing", which is often described as a series of stationary frames of an object's image that trail behind it as it moves, somewhat akin to the visual effects of a strobe light.(Abraham, 1983; DSM-IV 1994; *WWWref?). Under the proposed hypothesis, this effect could be attributed to the asynchronous processing of end-stop and line-detector information. Presumably, in normal vision, these would be temporally correlated in some way that is meaningful to cells in higher levels of processing, which would construct our fine-grained perception of object edges from some synthesis of information from both edge-selective and end-stopped cells coding for specific contours. The unnatural inhibition caused by LSD in visual areas, namely the claustrum and V1, may alter this temporal structure in a way that dissociates the processing, and therefore our perception of these two sources of contour information. Perhaps, then, "trails" are the product of lagged and uncoordinated processing of end-inhibition information relative to other sources of edge information. Since the claustral target zone for V1 is thought to overlap somewhat with those of higher-level areas involved in motion processing and eye movements, normally the claustrum may use signals relating to the expected trajectory of a moving object or imminent eye movements to update its output to V1. Thus, inhibition of claustral projecting cells may eliminate this predictive modulation of contour processing, which could also give rise to trailing effects.
Some evidence for LSD's effects on the temporal aspects of vision come from a study that measured critical flicker frequency in LSD users and control subjects (Abraham and Wolf 1988). This test presents subjects with a rapidly flickering visual stimulus and measures the highest frequency at which they can still perceive the flicker (before it is fused by our brain into a single image, just the like the image on a computer monitor). LSD users were found to require slower frequencies to perceive the flicker, especially in the peripheral visual field.
Similar mechanisms may be at work in false perceptions of object movement in the visual periphery, a commonly reported acute effect of LSD and symptom of HPPD (Abraham, 1983; DSM-IV 1994; *WWWref?). Sudden influxes of hallucinogen into visual areas of the brain, or perhaps serotonin in HPPD, may rapidly uncouple different sources of information about the edges of objects, which could produce the perceptual consequences of "a stable object appear[ing] to wave, roll, or jump" (Abraham, 1983). Since the visuotopic map of the claustrum emphasizes the peripheral field representation more than most cortical visual areas, and its projection to V1 preferentially terminates in its representation of the periphery (LeVay and Sherk 1981b), the claustrum seems a likely locus of these effects.
Many LSD users and sufferers of HPPD also report seeing "halos" or "mists" surrounding objects, or the perception of objects as either larger (macropsia) or smaller (micropsia) than they really are (Abraham 1983). Halos may represent a minor form of the uncoupling of different sources of edge information. Perhaps, due to inhibition in the claustrum and V1, neurons in layer 4 of V1 develop abnormal end-stopping properties such that some subpopulations erroneously signal for a border that surrounds the true contours of an object, producing a halo, while most cells still correctly signal these contours. Macropsia and micropsia may be extreme forms of this putative halo-inducing mechanism, where edge detection mechanisms become dominated by input from these incorrectly coding cells, causing the entire perceived object borders to be extended out or withdrawn in from their correct positions in space.
A somewhat rarer hallucinatory phenomenon produced by LSD (but not seen in HPPD) is synesthesia, the experiencing of a stimulus of one sensory modality as if it were in a different modality. A commonly reported example of synesthesia is "seeing music" as shapes or colors (***). Anatomical studies have shown that, in the macaque monkey, the claustral connection zones of area 22, a higher order auditory cortical area, show some overlap with those of V4, IT, and MT, higher-order visual cortical areas thought to be involved in the processing of form and color, object perception, and motion, respectively (Sherk 1986; Boussaoud et al. 1992). This suggests that perhaps the claustrum has some normal cross-modal function, such as inhibiting the inappropriate activation of one modality when another is activated or attended to, and that this mechanism is inhibited or altered significantly by psychedelics to produce synesthesia.
Carefully planned and controlled psychophysical studies of the distorting effects of LSD and other hallucinogens on vision, of which there are few in the literature, could be an interesting and informative tool to aid in our understanding of the neural substrates of vision and consciousness. Unfortunately, strict regulations on subject use and even more severe restrictions on the use of psychedelic drugs in research make it very difficult to test the acute effects of these substances on vision. However, the persistence some of these effects in a small population of former users allows researchers a close approximation to the acute effects, without the confounds of the cognitive distortions they often produce. A few studies have examined visual function in persons with HPPD and former LSD users, with fairly interesting results: these subjects showed abnormality in a test of color perception (Abraham 1982) and higher thresholds for perceiving flicker and for detecting a stimulus while adapting to a dark environment (Abraham and Wolf 1988). However, extensive, detailed psychophysical studies that specifically address the types of distortions seen in HPPD have yet to be undertaken.
One interesting avenue of research is suggested by a study that found long-term deficits in a visual size discrimination task in monkeys after they were administered LSD (Sharpe et al. 1967). A similar test of the discrimination of the size of objects or the length of lines in humans with HPPD could be an effective probe into the mechanisms that produce micropsia, macropsia, and "halos". If psychedelic visual distortions are, as hypothesized, mediated by a dysfunction of cortical mechanisms that detect object contours based on end-stopping information, then a direct assay of the perception of illusory contours, thought to be supported mainly by end-stopped mechanisms, is in order. An effective analytical tool used in previous studies of illusory contour perception (in normal subjects) is the discrimination of the orientation of an illusory border (Westheimer and Li 1996). Another possible test is the discrimination of the direction of motion of an illusory contour, which could uncover an exacerbation of these hypothesized visual deficits by moving stimuli.
To directly investigate the distortions of motion perception caused psychedelics, it would interesting to compare their perceptions of real and apparent motion (illusions of motion induced by an object simultaneously disappearing from the visual field and "reappearing" elsewhere). As the perception of apparent motion is thought to be mediated by higher-level attention-based mechanisms, the presence or absence of "trails" in apparent motion would give an indication of whether trails are caused high or low-level motion detection mechanisms. Another possibly informative test would be to probe the detection of brief stimuli presented at various spatiotemporal positions surrounding a moving object (real motion). Elevated or depressed detection thresholds in HPPD could uncover the spatial and temporal properties and nature of the trail-inducing mechanism. Finally, based on previous empirical evidence (Abraham and Wolf 1988) and the proposed hypothesis, it might be informative to compare performance in the central and peripheral visual fields on most of these tests.
It is obvious that a full explanation of the perceptual and cognitive distortions caused by psychedelics is considerably beyond our current understanding of their action and the neural circuits they interact. Furthermore, even with the small body of knowledge we do have on this subject, it is evident that psychedelics interact with a number of areas of the brain not covered in this review, many of which probably contribute to their hallucinatory effects. However, I have demonstrated in this paper that fairly simple mechanisms can be posited that explain a number of the characteristic perceptual distortions caused by these drugs. More importantly, this and other hypotheses can be fairly easily put to test on persons with hallucinogen persisting perception disorder, which could lead to a better understanding of the mechanisms of action of psychedelics, new methods of treatment of HPPD, and possibly to important insights into normal visual function.
Abeles, M. Corticonics : neural circuits of the cerebral cortex. New York : Cambridge University Press, 1991.
Abi-Saab, W.M., Grobin, A.C., Cubells, J.F., and Deutch, A.Y. Serotonin 5-HT-2 receptor-mediated modulation of GABA neurons in the rat prefrontal cortex. Society for Neuroscience Abstracts 22: 1297, 1996.
Abraham, H.D. A chronic impairment of colour vision in users of LSD. British Journal of Psychiatry 140: 518-520, 1982.
Abraham, HD. Visual phenomenology of the LSD flashback. Archives of General Psychiatry 40: 884-889, 1983.
Abraham, H.D. and Wolf, E. Visual function in past users of LSD: psychophysical findings. Journal of Abnormal Psychology 97: 443-447, 1988.
Aghajanian, G.K. Serotonin and the action of LSD in the brain. Psychiatric Annals 24: 137-41, 1994.
Baizer, J.S., Desimone, R., and Ungerleider, L.G. Comparison of Subcortical Connections of Inferior Temporal and Posterior Parietal Cortex in Monkeys. Visual Neuroscience 10: 59-72, 1993.
Bolz, J. and Gilbert, C.D. Generation of end-inhibition in the visual cortex via interlaminar connections. Nature 320: 362-365, 1986.
Boussaoud, D., Desimone, R., and Ungerleider, L.G. Subcortical connections of visual areas MST and FST in macaques. Visual Neuroscience 9: 291-302, 1992.
Diagnostic and statistical manual of mental disorders : DSM-IV. 4th ed. Washington, DC : American Psychiatric Association, 1994.
Foote, W.E. Electrophysiological studies of d-lysergic acid diethylamide in the visual system. Neuroscience and Biobehavioral Reviews 6: 503-507, 1982.
Godbout, R., Mantz, J., Glowinski, J., and Thierry, A-M. The novel 5-HT2 receptor antagonist, RP 62203, selectively blocks serotoninergic but not dopaminergic-induced inhibition in the rat prefrontal cortex. European Journal of Pharmacology 204: 97-100, 1991.
LeVay, S. and Sherk, H. The visual claustrum of the cat. I. Structure and connections. Journal of Neuroscience 1: 956-980, 1981a.
LeVay, S. and Sherk, H. The visual claustrum of the cat. II. The visual field map. Journal of Neuroscience 1: 981-992, 1981b.
LeVay, S. Synaptic organization of claustral and geniculate afferents to the visual cortex of the cat. Journal of Neuroscience 6: 3564-3575, 1986.
Lidow, M.S., Goldman-Rakic, P.S., Gallagher, D.W., and Rakic, P. Quantitative autoradiographic mapping of serotonin 5-HT1 and 5-HT2 receptors and uptake sites in the neocortex of the rhesus monkey. Journal of Comparative Neurology 280: 27-42, 1989.
Marek, G. J. and Aghajanian, G. K. LSD and the phenethylamine hallucinogen DOI are potent partial agonists at 5-HT-2A receptors in interneurons in rat piriform cortex. Journal of Pharmacology and Experimental Therapeutics 278: 1373-1382, 1996.
Mckenna, D.J., Nazarali, A.J., Hoffman, A.J., Nichols, D.E., Mathis, C.A., and Saavedra, J.M. Common receptors for hallucinogens in rat brain: A comparative autoradiographic study using iodine-125 LSD and iodine-125 DOI, a new psychotomimetic radioligand. Brain Research 476: 45-56, 1989.
Rasmussen, K. and Aghajanian, G.K. Effects of hallucinogens on spontaneuous and sensory-evoked locus coeruleus activity in the rat: reversal by selective 5-HT-2 antagonists. Brain Research 385: 395-400, 1986.
Saavedra, J.M. Common receptors for hallucinogens in rat brain: A comparative autoradiographic study using iodine-125 LSD and iodine-125 DOI, a new psychotomimetic radioligand. Brain Research 476: 45-56, 1989.
Selemon, L.D. and Goldman-Rakic, P.S. Common cortical and subcortical targets of the dorsolateral prefrontal and posterior parietal cortices in the rhesus monkey: Evidence for a distributed neural network subserving spatially guided behavior. Journal of Neuroscience 8: 4049-4068, 1988.
Sharpe, L.G., Otis, L.S., and Schusterman, R.J. Disruption of size discrimination in squirrel monkeys (Saimiri sciureus) by LSD-25. Psychonomic Science 7: 103-104, 1967.
Sheldon, P.W. and Aghajanian, G.K. Serotonin (5-HT) induces IPSPs in pyramidal layer of rat piriform cortex: evidence for the involvement of a 5-HT-2-activated interneuron. Brain Research 506: 62-69, 1990.
Sherk, H. and LeVay, S. The visual claustrum of the cat. III. Receptive field properties. Journal of Neuroscience 1: 993-1002, 1981.
Sherk H. and LeVay, S. Contribution of the cortico-claustral loop to receptive field properties in area 17 of the cat. Journal of Neuroscience 3: 2121-2127, 1983.
Sherk, H. The claustrum and the cerebral cortex. In Cerebral Cortex, Vol. 5. ed. Jones, E.G. and Peters, A. New York: Plenum Press, 1986.
Singer, W. Neurophysiology: the changing face of inhibition. Current Biology 6: 395-397, 1996.
Smiley, J.F. and Goldman-Rakic, P.S. Serotonergic axons in monkey prefrontal cerebral cortex synapse predominantly on interneurons as demonstrated by serial section electron microscopy. Journal of Comparative Neurology 367: 431-443, 1996.
Titeler, M., Lyon, R.A., and Glennon, R.A. Psychopharmacology 94: 213-216, 1988.
von der Heydt R. and Peterhans E. Mechanisms of contour perception in monkey visual cortex. I. Lines of pattern discontinuity. Journal of Neuroscience 9: 1731-1748, 1989.
Watts, S.W., Gackenheimer, S.L., Gehlert, D.R., and Cohen, M.L. Autoradiographic comparison of (125I)LSD-labeled 5-HT-2A receptor distribution in rat and guinea pig brain. Neurochemistry International 24: 565-574, 1994.
Westheimer, G. and Li, W. Classifying illusory contours by means of orientation discrimination. Journal of Neurophysiology 75: 523-528, 1996.